network connection to agent should be well secured by default
design doc: https://gitlab.isc.org/isc-projects/stork/-/wikis/Designs/Agent-Server-Security
There are a couple of aspects to be addressed here:
-
confidentiality -
mutual authentication -
authorization
This work has been split to separate issues:
- #479 (closed) (keys and certs generation)
- #481 (closed) (agent registration in the server)
- #483 (closed) (agent deployment and authorization UI)
- #485 (closed) (demo update)
- #486 (closed) (docs)
- #507 (closed) (system tests)